Our long-term goal is to understand the neural mechanisms underlying complex sound recognition, the formation of perceptual acoustical memories and the role of these memories in guiding vocal learning. In order to achieve our goal, we are studying neural processing in the auditory midbrain and forebrain of songbirds in the context of their song learning behavior. Songbirds learn to produce their complex songs by comparing the auditory feedback from their own vocalizations with a memorized version of a tutor's song. This memory based vocal learning is reminiscent of some aspects of speech learning in humans and very rare in non-avian animals. We have shown that auditory neurons in the forebrain of songbirds are on average more responsive to the songs produced by the species than to matched synthetic sounds. We will obtain the stimulus-response function of auditory neurons at different stages of the auditory processing stream to understand how this auditory selectivity is generated in the auditory system. In particular, we will estimate the spectro-temporal receptive fields of auditory neurons and compare them to the spectro- temporal amplitude modulations found in song and in other natural sounds. We will also investigate the effect of early sensory experience on this acoustical processing. We will then investigate the connections between the auditory system and the vocal system responsible for song production and learning in male songbirds. We will also compare the auditory system of male songbirds with that of female songbirds. It is known that the song system is sexually dimorphic in songbirds and we will investigate the extent to which this sexual dimorphism affects the perceptual system. Our results will provide insights into the mechanisms underlying complex sound recognition and production in humans, in particular as they relate to speech comprehension and learning. [unreadable] [unreadable]